201212595 六、發明說明： 【發明所屬之技術領域】 本發明係關於將IQ基頻訊號作截波處理的峰値電力 抑制電路、及具此電路的通信裝置。更具體而言，關於用 以更適當地進行輸入至無線送訊機中的電力放大電路的IQ 基頻訊號的振幅限制的截波方法的改良。 【先前技術】 例 如 OFDM(Orthogonal Frequency Division Multiplex:正交分頻多工）方式或 W-CDMA(Wideband Code Division Multiple Access)方式般使用複數載波來調變送訊 訊號的方式有時載波的相位相疊合而形成具有較大峰値電 力的訊號。 另一方面，電力放大器要求優異線形性，但若輸入超 出最大輸出之位準的訊號，則輸出會飽和而非線性失真會 增大。 爲此，若較大峰値電力之訊號輸入非線性放大器，則 輸出訊號發生非線性失真，成爲收訊側的收訊特性劣化或 頻帶外幅射的原因。 爲了對峰値電力不會使非線性失真增大，動態範圍較 大的電力放大器成爲必要，但若爲了不會頻繁出現的峰値 電力而加寬放大器的動態範圍，則時間軸上之波形的平均 電力與短時間的峰値電力的比（PAPR: Peak to Average Power Ratio)會變大，電力效率會變差》 201212595 因此，關於出現頻度較低之大峰値電力的訊號，在輸 入前進行抑制較照原樣輸入至放大器合理。於是，爲了抑 制電力放大前的IQ基頻訊號的峰値電力，往往對超過預定 臨限値的峰値電力的IQ基頻訊號進行在瞬間供予逆向振 幅的截波處理。 該截波處理係在時間軸上逆向施加脈衝狀訊號的處 理，因此同在頻率軸上施加寬頻帶的雜訊。因此，僅單純 進行截波處理時，會有在頻帶外產生雜訊的問題。 因此，爲了應付該頻帶外輻射的問題，已知被稱爲 NS-CFR(Noise Shaping-Crest Factor Reduction)及 PC-CFR(Peak Cancellation-Crest Factor Reduction)的峰値 電力抑制電路。 其中，NS-CFR電路係對瞬時電力超過臨限値的IQ基 頻訊號的峰値成分（臨限値的增量），以低通濾波器或 FIR(Finite Impulse Response)濾波器等進行瀘波來作頻帶 限制，由原本的IQ基頻訊號減去該頻帶限制後的峰値成分 (參照專利文獻1)。 此外，PC-CFR電路係預先設定用途爲即使作截波亦不 會使頻帶外輻射產生的抵消用脈衝（基本函數波形），由原 本的IQ基頻訊號減去對瞬時電力超過臨限値的IQ基頻訊 號的峰値成分（臨限値的增量）乘上該抵消用脈衝所求出的 抵消訊號（參照專利文獻2及3)。 (先前技術文獻） 201212595 (專利文獻） 專利文獻1 :專利第3 95434 1號公報 專利文獻‘2:專利第3853509號公報 專利文獻3:特開2004-135087號公報（第1圖〜第6 圖） 【發明內容】 (發明所欲解決之課題） 總言之，上述CFR電路之截波處理的本質係在超過臨 限値的峰値電力的發生瞬間，對IQ基頻訊號施加抵消訊 號，以將峰値壓制在該臨限値程度。 爲此，以抵消用脈衝與增量相乘所得的抵消訊號來抵 消峰値的PC-CFR電路係抵消用脈衝的脈衝寬度愈窄，愈 可抵消目標的瞬間峰値，可進行對之後的訊號波形不會造 成影響的理想截波處理。 但是，習知的PC-CFR電路係利用送訊時所使用的頻 帶內的訊號成分來生成抵消用脈衝，因此並無法使該脈衝 寬度太細。 因此，若以使用上述抵消用脈衝的抵消訊號來抵消IQ 基頻訊號時，則抵消訊號會干擾峰値時之後的訊號波形而 生成新的峰値波形，會有無法確實地抑制IQ基頻訊號之峰 値電力的情形。 本發明係鑑於該習知的問題點，目的在提供一種可更 加確實地抑制IQ基頻訊號之峰値電力的峰値電力抑制電 201212595 路等。. (解決課題之手段） (1) 本發明之峰値電力抑制電路係將IQ基頻訊號作截 波處理的峰値電力抑制電路，其特徵爲具備：電力計算部， 係計算前述IQ基頻訊號的瞬時電力；脈衝保持部，係保持 具有前述IQ基頻訊號的頻帶內及該頻寬外雙方之頻率成 分的抵消用脈衝；及截波處理部，係對所被計算出的前述 瞬時電力大於預定臨限値的前述IQ基頻訊號，減去對由該 訊號的前述臨限値的增量乘上前述抵消用脈衝所得的抵消 訊號。 根據本發明之峰値電力抑制電路，上述截波處理部從 該IQ基頻訊號減去一抵消訊號，而該抵消訊號係對由IQ 基頻訊號的臨限値的增量乘上不僅具有IQ基頻訊號的頻 帶內的頻率成分而亦具有其頻寬外的頻率成分的抵消用脈 衝而得到。 爲此，具有頻寬外的頻率成分的上述抵消訊號係成爲 可在短時間內改變的尖銳脈衝狀的訊號，將其減去可防止 產生新的峰値波形，並可更加確實地抑制IQ基頻訊號的峰 値電力。 (2) 在本發明之峰値電力抑制電路中，具體而言，前述 抵消用脈衝可由一合成脈衝所構成，而該合成脈衝係具有 頻寬內的頻率成分且主波瓣區間的能量局部率爲85〜99% 的基本脈衝與具有在該基本脈衝的峰値保持的時間內急遽 201212595 上升的頻寬外成分且寬度比前述基本脈衝更細且峰値位準 較低的輔助脈衝合成之脈衝。 (3) 但是，上述輔助脈衝的頻率成分係遍及包含IQ基 頻訊號的頻帶的廣範圍之頻帶，因此若由IQ基頻訊號減去 使用包含該輔助脈衝的合成脈衝所求出的抵消訊號，會成 爲與遍及廣頻帶被施加雜訊的情形相同，因此若未適當設 定基本脈衝與輔助脈衝的位準，則會有頻帶內的通信品質 (EVM: Error Vector Magnitude)降低、或因對頻寬外的洩 漏電力而發生不需要的雜訊等可能性。 因此，較佳爲在本發明之峰値電力抑制電路中，前述 基本脈衝與前述輔助脈衝的峰値位準以在前述IQ基頻訊 號的頻帶內滿足所希望的EVM(Error Vector Magnitude)、 且滿足所希望的鄰接通道洩漏電力比（ACLR : Adjacent Channel Leakage Ratio)的方式予以設定。 (4) 更具體而言，較佳爲當將前述基本脈衝的峰値位準 設爲α、將前述輔助脈衝的峰値位準設爲β時，以滿足 0.03$β/α$0.1的方式，設定該等位準α，β的比率。 其理由如以下之實施形態中亦會詳述，β/α = 0.1係同 時滿足平常所要求的EVM與ACLR的最大比率，因此若將 該比率β/α的輔助脈衝與基本脈衝合成，可得最大峰値抑 制效果。 此外，設爲β/α20·03係因爲若該比率β/α未達〇.〇3, 則輔助脈衝的位準會變得過小，會有無法適當消除抵消訊 201212595 號的減算所產生的新峰値波形之虞。 (5) 較佳爲本發明之峰値電力抑制電路另外具備一臨 限値更新部，而該臨限値更新部係在每一個前述1Q基頻訊 號的平均電力有時間上變動的可能性的控制周期，更新在 前述截波處理部所使用的前述臨限値。 此時，上述臨限値更新部係在每一個上述控制周期’ 更新在截波處理部所使用的臨限値，因此即使爲例如1Q基 頻訊號的平均電力較小的時間帶，亦可確實地抑制瞬時電 力。 (6) 此外，較佳爲本發明之峰値電力抑制電路另外具備 —脈衝生成部，而該脈衝生成部係以可將前述IQ基頻訊號 對應每一個其頻帶的平均電力地抵消的方式生成前述抵消 用脈衝。 此時，使用在上述脈衝生成部所生成的抵消用脈衝使 得不會有平均電力較小者的頻寬的送訊電力因抵消訊號的 減算而降低所需以上的情形，即使是平均電力按每一個頻 帶而異的IQ基頻訊號的情形，亦可在不使SNR降低之情 況下適當地進行截波處理。 (7) 本發明之通信裝置係具有一送訊機，該送訊機搭載 有本發明之峰値電力抑制電路、及配置在其後段的電力放 大電路，該通信裝置達成與本發明之峰値電力抑制電路相 同的作用效果。 (發明之效果） 201212595 如以上所示，本發明使用不僅具有IQ基頻訊 內的頻率成分而亦具有該頻寬外的頻率成分的抵 來進行截波處理，因此可更加確實地抑制IQ基頻 値電力。 【實施方式】 以下一面參照圖式，一面說明本發明之實施 〔第1實施形態〕 〔無線通信系統〕 第1圖係適於適用本發明之第1實施形態之 系統的全體構成圖。 如第1圖所示，本實施形態之無線通信系統 台裝置（BS : Base Station)〗、及在該裝置1的單5 與該裝置1進行無線通信的複數移動終端機（MS Station)2 所構成。 此無線通信系統係採用OFDM方式作爲基地 與移動終端機2之間的調變方式。該方式係使送 置於多數載波（副載波）的多載波數位調變方式， 彼此正交，因此具有在頻率軸重疊發生越多資料 排列的優點。 ith外’本實施形態之無線通信系統係由適用 EV〇iution)方式的行動電話用系統所構成， 台裝置1與移動終端機2之間進行依據LTE方式 其中’可適用本發明之無線通信系統並非侷201212595 SUMMARY OF THE INVENTION [Technical Field] The present invention relates to a peak-to-peak power suppression circuit for intercepting an IQ fundamental frequency signal, and a communication device having the same. More specifically, it relates to an improvement of a clipping method for more appropriately performing amplitude limitation of an IQ fundamental signal input to a power amplifying circuit in a wireless transmitter. [Prior Art] For example, the OFDM (Orthogonal Frequency Division Multiplex) method or the W-CDMA (Wideband Code Division Multiple Access) method uses a complex carrier to modulate the transmission signal. Superimposed to form a signal with a large peak power. On the other hand, power amplifiers require excellent linearity, but if a signal that exceeds the level of the maximum output is input, the output will be saturated and the nonlinear distortion will increase. For this reason, if the signal of the larger peak power is input to the non-linear amplifier, the output signal is nonlinearly distorted, which causes the reception characteristics of the receiving side to deteriorate or the out-of-band radiation. In order to increase the nonlinear distortion of the peak-to-peak power, a power amplifier having a large dynamic range is necessary, but if the dynamic range of the amplifier is widened in order to prevent the peak-to-peak power from occurring frequently, the waveform on the time axis is The ratio of average power to short-term peak-to-peak power ratio (PAPR: Peak to Average Power Ratio) will increase, and power efficiency will deteriorate. 201212595 Therefore, the signal of the peak frequency and power with low frequency is suppressed before input. It is reasonable to input to the amplifier as it is. Therefore, in order to suppress the peak-to-peak power of the IQ fundamental frequency signal before the power amplification, the IQ fundamental signal of the peak-to-peak power exceeding the predetermined threshold is often subjected to the clipping process of the reverse amplitude in an instant. The chopping process is a process of applying a pulse signal in the reverse direction on the time axis, thus applying a broadband noise on the frequency axis. Therefore, when the chopping process is simply performed, there is a problem that noise is generated outside the band. Therefore, in order to cope with the problem of the out-of-band radiation, a peak-to-peak power suppression circuit called NS-CFR (Noise Shaping-Crest Factor Reduction) and PC-CFR (Peak Cancellation-Crest Factor Reduction) is known. Among them, the NS-CFR circuit performs chopping with a low-pass filter or a FIR (Finite Impulse Response) filter for the peak-to-peak component of the IQ fundamental frequency signal whose instantaneous power exceeds the threshold (. For the band limitation, the peak component after the band limitation is subtracted from the original IQ fundamental signal (see Patent Document 1). In addition, the PC-CFR circuit is pre-set to use a canceling pulse (basic function waveform) that does not cause out-of-band radiation even if it is cut off, and subtracts the instantaneous power from the original threshold by the original IQ fundamental signal. The peak component of the IQ fundamental frequency signal (increase in threshold 乘) is multiplied by the cancellation signal obtained by the cancellation pulse (see Patent Documents 2 and 3). (Prior Art Document) 201212595 (Patent Document) Patent Document 1: Patent No. 3,954,434, Patent Document No. 3, Patent No. 3,853,509, Patent Document 3: JP-A-2004-135087 (Fig. 1 to Figure 6) SUMMARY OF THE INVENTION [Problems to be Solved by the Invention] In summary, the essence of the chopping process of the CFR circuit is to apply a cancellation signal to the IQ fundamental frequency signal at the moment when the peak power of the threshold voltage exceeds the threshold. The peak is suppressed to the extent of the threshold. For this reason, the PC-CFR circuit that cancels the peak by the offset signal multiplied by the pulse and the increment is offset, and the pulse width of the canceling pulse is narrower, and the instantaneous peak of the target can be cancelled, and the subsequent signal can be performed. The ideal chopping process that the waveform does not affect. However, the conventional PC-CFR circuit generates a canceling pulse by using a signal component in a frequency band used for transmission, and therefore the pulse width cannot be made too thin. Therefore, if the IQ fundamental signal is cancelled by using the cancellation signal of the cancellation pulse, the cancellation signal will interfere with the signal waveform after the peak time to generate a new peak waveform, and the IQ fundamental signal cannot be reliably suppressed. The peak of the power situation. The present invention has been made in view of the conventional problems, and an object of the invention is to provide a peak power reduction circuit 201212595 which can more reliably suppress peak-to-peak power of an IQ fundamental frequency signal. (Means for Solving the Problem) (1) The peak power suppression circuit of the present invention is a peak power suppression circuit that performs a chirp processing on an IQ fundamental signal, and is characterized in that it includes a power calculation unit that calculates the IQ fundamental frequency. The instantaneous power of the signal; the pulse holding unit is a canceling pulse for maintaining a frequency component of both the frequency band of the IQ fundamental frequency signal and the outside of the bandwidth; and a clipping processing unit for calculating the instantaneous power calculated The aforementioned IQ fundamental frequency signal greater than the predetermined threshold 减 is subtracted from the offset signal obtained by multiplying the aforementioned offset pulse by the aforementioned offset pulse. According to the peak power suppression circuit of the present invention, the clipping processing unit subtracts a cancellation signal from the IQ fundamental signal, and the cancellation signal multiplies the increment of the threshold of the IQ fundamental signal by not only IQ. The frequency component in the frequency band of the fundamental frequency signal is also obtained by canceling the pulse of the frequency component outside the bandwidth. For this reason, the above-mentioned cancellation signal having a frequency component outside the bandwidth becomes a sharp pulse-like signal that can be changed in a short time, and subtracting it can prevent a new peak-to-peak waveform from being generated, and can more reliably suppress the IQ-based signal. The peak frequency of the frequency signal. (2) In the peak power suppression circuit of the present invention, specifically, the canceling pulse may be constituted by a combined pulse having a frequency component within a bandwidth and an energy local rate of the main lobe section a pulse of 85 to 99% of the basic pulse and the auxiliary pulse having the outer component of the bandwidth which is rushed by the peak of the peak of the basic pulse and which is increased by 201212595 and having a width smaller than the aforementioned basic pulse and having a lower peak level. . (3) However, the frequency component of the auxiliary pulse is in a wide frequency band of the frequency band including the IQ fundamental signal. Therefore, if the cancellation signal obtained by using the synthesized pulse including the auxiliary pulse is subtracted from the IQ fundamental signal, This is the same as the case where noise is applied to the wide frequency band. Therefore, if the level of the basic pulse and the auxiliary pulse are not properly set, the communication quality in the frequency band (EVM: Error Vector Magnitude) is lowered, or the bandwidth is affected. There is a possibility of leakage of electric power and unwanted noise. Therefore, in the peak power suppression circuit of the present invention, it is preferable that the peak pulse level of the basic pulse and the auxiliary pulse satisfy a desired EVM (Error Vector Magnitude) in a frequency band of the IQ fundamental frequency signal, and This is set by the way of satisfying the desired adjacent channel leakage power ratio (ACLR: Adjacent Channel Leakage Ratio). (4) More specifically, it is preferable to set the peak 値 level of the basic pulse to α and the peak 値 level of the auxiliary pulse to β to satisfy 0.03$β/α$0.1. The ratio of the levels α, β is set. The reason is also described in the following embodiments. β/α = 0.1 system satisfies the usual maximum ratio of EVM and ACLR, so if the auxiliary pulse of the ratio β/α is synthesized with the basic pulse, Maximum peak suppression effect. In addition, it is set to β/α20·03 because if the ratio β/α does not reach 〇.〇3, the level of the auxiliary pulse will become too small, and the new generation due to the subtraction of the 201212595 can not be properly eliminated. The peak of the peak waveform. (5) Preferably, the peak power suppression circuit of the present invention further includes a threshold update unit, and the threshold update unit has a possibility that the average power of each of the 1Q fundamental signals changes in time. The control period updates the aforementioned threshold used by the above-described intercept processing unit. In this case, the threshold update unit updates the threshold used by the occlusion processing unit in each of the control periods. Therefore, even if the average power of the 1Q fundamental frequency signal is small, for example, it can be confirmed. Ground power is suppressed. (6) Further, it is preferable that the peak power suppression circuit of the present invention further includes a pulse generation unit that generates the analog base frequency signal so as to cancel the average power of each of the frequency bands. The aforementioned cancellation pulse. In this case, the canceling pulse generated by the pulse generating unit is used so that the transmission power of the bandwidth having a smaller average power is not required to be reduced by the subtraction of the cancel signal, even if the average power is per In the case of a frequency-dependent IQ fundamental signal, it is also possible to appropriately perform the chopping process without lowering the SNR. (7) The communication device of the present invention includes a transmitter equipped with the peak power suppression circuit of the present invention and a power amplifier circuit disposed in the subsequent stage, and the communication device achieves the peak of the present invention. The same effect of the power suppression circuit. (Effects of the Invention) As described above, the present invention uses the frequency component of the IQ-based frequency band and the frequency component outside the bandwidth to perform the chopping process, thereby suppressing the IQ basis more reliably. Frequent power. [Embodiment] The present invention will be described with reference to the drawings. [First Embodiment] [Wireless communication system] Fig. 1 is a view showing an overall configuration of a system to which the first embodiment of the present invention is applied. As shown in Fig. 1, a wireless communication system station device (BS: Base Station) of the present embodiment and a plurality of mobile terminal devices (MS Station) 2 that wirelessly communicate with the device 1 in the device 5 of the device 1 Composition. This wireless communication system uses the OFDM method as a modulation method between the base and the mobile terminal 2. This method has a multi-carrier digital modulation method that is transmitted to a plurality of carriers (subcarriers) and is orthogonal to each other, so that there is an advantage that data arrangement occurs more frequently in the frequency axis overlap. In addition, the wireless communication system of the present embodiment is constituted by a mobile phone system in which the EV〇iution method is applied, and the wireless communication system in which the present invention is applicable according to the LTE method is performed between the station device 1 and the mobile terminal device 2. Not a bureau
號的頻帶 消用脈衝 訊號的峰 形態。 無線通信 係由基地 (cell)內 ：Mobile 台裝置1 訊資料載 各副載波 越稠密地 LT E(L ο n g 在各基地 的通信。 限於LTE -10- 201212595 方式者，亦可爲w-CDMA方式，以下假定本發明應用於LTE 方式之基地台裝置1的情形來進行說明。 〔LTE的下行訊框〕 第6圖係顯示LTE的下行訊框的構造圖。圖中，縱軸 方向係表示頻率，橫軸方向係表示時間。 如第6圖所示，構成下行（DL)訊框的合計1〇個子訊框 (subframe#0〜#9)係分別由2個時槽（slot#0與slot#l)所構 成，1個時槽係由7個OFDM符號所構成（Normal Cyclic Prefix的情形）。 此外，圖中屬於資料傳送方面之基本單位的資源區塊 (RB: Resource Block)係在頻率軸方向設定爲12副載波， 並在時間軸方向'設定爲70FDM符號（1時槽）。 因此，例如DL訊框的頻帶被設定爲5MHz時係配列有 3 00個副載波，因此資源區塊係朝頻率軸方向配置25個》 其中，1個子訊框的送訊時間爲1 ms，本實施形態構成 1個子訊框的2個時槽分別包含7個OFDM符號，因此1 個 OFDM符號的送訊周期（符號周期）爲 l/14ms(=約 0.071ms)。 如第6圖所示，在各子訊框的前頭分配一控制通道， 而該控制通道用於供基地台裝置1對移動終端機2傳送下 行通信所需的資訊。 該控制通道係儲存D L控制資訊 '該子訊框的資源分 配資訊、混合式自動重送請求（HARQ: Hybrid Automatic 201212595The band of the number eliminates the peak form of the pulse signal. The wireless communication system is in the cell (cell): the mobile station device 1 data carries the sub-carriers more densely LT E (L ο ng communication in each base. Limited to LTE -10- 201212595 mode, can also be w-CDMA In the following, a case where the present invention is applied to the LTE-based base station apparatus 1 will be described. [LTE downlink frame] FIG. 6 is a structural diagram showing a downlink frame of LTE. In the figure, the vertical axis direction indicates The frequency and the horizontal axis indicate the time. As shown in Fig. 6, the total of one sub-frames (subframe#0~#9) constituting the downlink (DL) frame are respectively composed of two time slots (slot#0 and In the case of slot #1), one time slot is composed of seven OFDM symbols (in the case of Normal Cyclic Prefix). In addition, the resource block (RB: Resource Block) belonging to the basic unit of data transmission in the figure is The frequency axis direction is set to 12 subcarriers, and is set to 70 FDM symbols (1 time slot) in the time axis direction. Therefore, for example, when the frequency band of the DL frame is set to 5 MHz, there are 300 subcarriers, so the resource area Block system is arranged in the direction of the frequency axis 25", one sub-frame The transmission time is 1 ms. In this embodiment, the two time slots constituting one subframe include seven OFDM symbols, so the transmission period (symbol period) of one OFDM symbol is l/14 ms (= about 0.071 ms). As shown in Fig. 6, a control channel is allocated at the head of each subframe, and the control channel is used for the base station device 1 to transmit information required for downlink communication to the mobile terminal 2. The control channel is stored. DL control information 'The resource allocation information of this sub-frame, hybrid automatic repeat request (HARQ: Hybrid Automatic 201212595
Report Request) 之收訊成功通知 （ACK ： Acknowledgement)、收訊失敗通知（N A C K : N e g a t i v e Acknowledgement)等 〇 在第6圖所示之DL訊框中，PBCH(PhysicalBroadcast CHannel)係用於藉廣播送訊而對終端機裝置通知系統的頻 寬等的廣播通道，第〇個（#〇)及第6個（#5)子訊框分配有用 於識別基地台裝置1或單元的訊號，即第1同步訊號 (P-SCH: Primary Synchronization CHannel)及第 2 同步訊 號（S-SCH : S econd ar y S y n chro ni z ati on C H annel) 〇 此外，未分配有上述各通道的其他區域（第6圖中沒有 影線的區域）的資源區塊係用作儲存使用者資料等用的DL 共有通信通道（PDSCH : Physical Downlink Shared CHannel)。 關於被儲存在上述PDSCH的使用者資料的分配，以被 分配在各子訊框前頭的上述控制通道內的資源分配資訊加 以規定，移動終端機2係可根據該資源分配資訊來判斷對 自己的資料是否已儲存在子訊框內。 〔送訊機的構成〕 第2圖係顯示基地台裝置1的OFDM送訊機3的主要 部分的功能方塊圖。 該送訊機3係具備送訊用處理器4與電力放大電路5, 送訊用處理器4係由例如在內部具有1或複數記憶體或 CPU 的 FPGA(FieId Programmable Gate Array)所構成。 -12- 201212595 上述FPG A係可在處理器出貨時或基地台裝置1製造 時等’預先設定各種邏輯電路的構成資訊（configuration)， 經由相關設定作業構成第2圖所示之各功能部6〜1 0。 亦即，本實施形態之送訊用處理器4係由左依序包 含· S/P 轉換部 6、映射部 7、IFFT(Inverse Fast Fourier Transform :高速傅立葉逆轉換）部8、訊號處理部9及正交 調變部10。 被輸入至送訊用處理器4的串列訊號列係在S /P (串列 並列）轉換部6中被轉換成複數訊號列，經轉換後的各並列 訊號列係在映射部7中被轉換成由預定振幅與相位的組合 所成的複數副載波訊號f 1、f2、... fn。 該各副載波訊號Π、f2、... fn係由IF FT部8轉換成在 時間軸上彼此正交的基頻訊號I訊號及Q訊號。 該IQ訊號（Iin，Qin)係在後段的訊號處理部（本實施形 態之峰値電力抑制電路）9中被施行預定的訊號處理。該訊 號處理後的IQ訊號（Iout’Qout)係在正交調變部1〇中被正 交調變而成爲調變波訊號，該調變波訊號係被輸入至後段 的電力放大電路5。 其中，本實施形態之峰値電力抑制電路9係以IQ基頻 訊號的瞬時電力P不會大於預定臨限値Pth的方式而將該 IQ基頻訊號進行截波處理者，詳細後述。 電力放大電路5係包含將由正交調變部10輸入的調變 波訊號轉換成類比訊號的D/A轉換電路、將轉換後的類比 -13- 201212595 訊號上轉換（up convert)成RF頻率的轉換器、及將該類比 訊號的電力放大的電力放大器，而放大後的RF訊號係從天 線送出外部。 以本實施形態之電力放大電路5而言，可爲電力放大 器的汲極電壓爲一定的固定電壓方式，而由達成高頻放大 器之高效率化的觀點來看，較佳爲採用 ET(Envelope Tracking)方式。 該ET方式的電力放大電路5係由輸入至電力放大器的 調變波訊號中抽出振幅資訊（波封），將與該振幅資訊相對 應的汲極電壓施加至電力放大器，使得電力放大器在接近 大致飽和的狀態下動作；同此，可減低固定電壓情形下動 作時所發生的電力損失，而可實現電力放大器的高效率化。 〔峰値電力抑制電路的構成〕 第3圖係本發明之第1實施形態之峰値電力抑制電路 9的功能方塊圖。 · 如第3圖所示，本實施形態之峰値電力抑制電路9係 包含電力計算部13、截波處理部17、及延遲部18,19。 其中，電力計算部13係算出由IQ基頻訊號的I成分 與Q成分的平方和所成的瞬時電力P。 本實施形態之截波處理部17係由一PC-CFR電路所構 成，而該PC-CFR電路係當IQ基頻訊號的瞬時電力P超過 預定的臨限値Pth時，由原本的IQ基頻訊號減去對由該臨 限値Pth的增量ΔΙ，Δ(^乘上預定的抵消用脈衝S所得的抵 14 - 201212595 消訊號Ic，Qc。 該截波處理部17係包含增量率計算部20、比較部21、 脈衝保持部22及加減法器23，24。 增量率計算部20係使用電力計算部1 3所算出的瞬時 電力P及預先設定的預定臨限値Pth而計算出相對於瞬時 電力P之臨限値Pth的增量率{ l-SQRT(Pth/P)}，並將該 增量率{ l-SQRT(Pth/P)}透過乘法器而與IQ基頻訊號的 各成分（I，Q)作乘算。 因此，超過IQ基頻訊號之臨限値Pth的部分的增量 △ I，AQ係根據下式算出。其中，此時，SQRT(.)係取括弧 內變數之平方根的函數（以下同）。 ΔΙ = { l-SQRT(Pth/P) } χΐReport Request) ACK: Acknowledgement, Negative Acknowledgement, etc. In the DL frame shown in Figure 6, PBCH (Physical Broadcast CHannel) is used to broadcast by broadcast. In response to the broadcast channel of the bandwidth of the terminal device notification system, the first (#〇) and the sixth (#5) subframes are assigned signals for identifying the base station device 1 or unit, that is, the first The synchronization signal (P-SCH: Primary Synchronization CHannel) and the second synchronization signal (S-SCH: S econd ar y Synchro ni z ati on CH annel) 〇 In addition, other areas of the above channels are not allocated (6th The resource block in the area where there is no hatching in the figure is used as a DL shared communication channel (PDSCH: Physical Downlink Shared CHannel) for storing user data and the like. The allocation of the user data stored in the PDSCH is defined by the resource allocation information allocated in the control channel at the head of each subframe, and the mobile terminal 2 can determine the self according to the resource allocation information. Whether the data has been stored in the sub-frame. [Configuration of Transmitter] Fig. 2 is a functional block diagram showing the main part of the OFDM transmitter 3 of the base station device 1. The transmitter 3 includes a transmission processor 4 and a power amplifier circuit 5. The transmitter processor 4 is composed of, for example, an FPGA (FieId Programmable Gate Array) having one or a plurality of memories or a CPU therein. -12- 201212595 The above FPG A system can set the configuration information of various logic circuits in advance when the processor is shipped or when the base station device 1 is manufactured, and the functional units shown in Fig. 2 are configured via the related setting operations. 6 to 1 0. In other words, the transmission processor 4 of the present embodiment includes the S/P conversion unit 6, the mapping unit 7, the IFFT (Inverse Fast Fourier Transform) unit 8, and the signal processing unit 9 in order from the left. And the quadrature modulation unit 10. The serial signal sequence input to the communication processor 4 is converted into a complex signal sequence in the S / P (serial parallel) conversion unit 6, and the converted parallel signal sequences are arranged in the mapping portion 7 Converted into complex subcarrier signals f 1 , f2, ... fn formed by a combination of predetermined amplitude and phase. The subcarrier signals Π, f2, ... fn are converted by the IF FT unit 8 into fundamental frequency signals I and Q signals which are orthogonal to each other on the time axis. The IQ signal (Iin, Qin) is subjected to predetermined signal processing in the signal processing unit (the peak power suppression circuit of the present embodiment) 9 in the subsequent stage. The IQ signal (Iout'Qout) after the signal processing is orthogonally modulated in the quadrature modulation unit 1A to become a modulated wave signal, and the modulated wave signal is input to the power amplifier circuit 5 of the subsequent stage. In the peak power suppression circuit 9 of the present embodiment, the IQ fundamental signal is subjected to the chopping process so that the instantaneous power P of the IQ fundamental frequency signal is not greater than the predetermined threshold 値Pth, which will be described in detail later. The power amplifying circuit 5 includes a D/A conversion circuit that converts the modulated wave signal input by the quadrature modulation unit 10 into an analog signal, and upconverts the converted analog-13-201212595 signal into an RF frequency. The converter and the power amplifier that amplifies the power of the analog signal, and the amplified RF signal is sent out from the antenna. In the power amplifier circuit 5 of the present embodiment, the drain voltage of the power amplifier can be a constant voltage method, and from the viewpoint of achieving high efficiency of the high-frequency amplifier, it is preferable to use ET (Envelope Tracking). )the way. The power amplifier circuit 5 of the ET system extracts amplitude information (wave seal) from a modulated wave signal input to the power amplifier, and applies a drain voltage corresponding to the amplitude information to the power amplifier so that the power amplifier is close to approximately In the saturated state, the power loss occurring during the operation under the fixed voltage can be reduced, and the efficiency of the power amplifier can be improved. [Configuration of the peak power suppression circuit] Fig. 3 is a functional block diagram of the peak power suppression circuit 9 according to the first embodiment of the present invention. As shown in Fig. 3, the peak power suppression circuit 9 of the present embodiment includes a power calculation unit 13, a wavelet processing unit 17, and delay units 18 and 19. The power calculation unit 13 calculates the instantaneous power P formed by the sum of the squares of the I component and the Q component of the IQ fundamental frequency signal. The chopping processing unit 17 of the present embodiment is constituted by a PC-CFR circuit which is based on the original IQ fundamental frequency when the instantaneous power P of the IQ fundamental frequency signal exceeds a predetermined threshold 値Pth. The signal subtracts the increment ΔΙ, Δ(^ multiplied by the predetermined offset pulse S from the threshold 値Pth by the offset 14 - 201212595, the cancellation signal Ic, Qc. The clipping processing unit 17 includes the increment rate calculation. The unit 20, the comparison unit 21, the pulse holding unit 22, and the adders/subtractors 23 and 24. The incremental rate calculating unit 20 calculates the instantaneous power P calculated by the power calculating unit 13 and a predetermined threshold 値Pth set in advance. Relative to the instantaneous power P, the increment rate lPth increment rate { l-SQRT(Pth/P)}, and the increment rate { l-SQRT(Pth/P)} is transmitted through the multiplier and the IQ fundamental frequency signal The components (I, Q) are multiplied. Therefore, the increment Δ I of the portion of the threshold 値Pth exceeding the IQ fundamental frequency signal, AQ is calculated according to the following equation. At this time, SQRT(.) is taken. The function of the square root of the variables in parentheses (the same below). ΔΙ = { l-SQRT(Pth/P) } χΐ
AQ = { 1 -SQRT(Pth/P) } xQ 比較部21係將在電力計算部13所算出的瞬時電力P 與臨限値Pth作比較，在瞬時電力p大於臨限値Pth時， 即對脈衝保持部22發出抵消用脈衝s的輸出指令。 脈衝保持部22係具有暫時保持由後述的合成脈衝所 構成的抵消用脈衝S(參照第4圖）之由雙埠RAM等所構成 的記憶體’當由比較部2〗接收到指令時，將所保持的抵消 用脈衝S乘上上述增量ai，Aq來算出抵消訊號ic，Qc。 此外’脈衝保持部22係在未由比較部21接收到指令 時’將上述增量M，Aq乘上零。 因此’關於瞬時電力P超過臨限値Pth的IQ基頻訊 -15- 201212595 號’根據下列數式所算出的抵消訊號Ic，Qc被輸入至加減 法器23,24 。AQ = { 1 - SQRT (Pth / P) } The xQ comparing unit 21 compares the instantaneous power P calculated by the power calculating unit 13 with the threshold 値 Pth, and when the instantaneous power p is greater than the threshold 値 Pth, The pulse holding unit 22 issues an output command for the canceling pulse s. The pulse holding unit 22 has a memory "composed of a double-pass RAM or the like" that temporarily holds the canceling pulse S (see FIG. 4) composed of a combined pulse to be described later, and when the command is received by the comparing unit 2, The canceled offset pulse S is multiplied by the above increments ai, Aq to calculate the canceling signals ic, Qc. Further, the pulse holding unit 22 multiplies the above-mentioned increments M, Aq by zero when the command is not received by the comparing unit 21. Therefore, the IQ signal -15-201212595 regarding the instantaneous power P exceeding the threshold 値Pth is input to the adder-subtracter 23, 24 according to the cancellation signal Ic, Qc calculated by the following equation.
Ic=AIxS= { 1-SQRT(Pth/P) } xIxSIc=AIxS= { 1-SQRT(Pth/P) } xIxS
Qc = AQ χ S = { 1 -SQRT(Pth/P) } xQxS 位於加減法器23,24之前段的延遲部18，19係使IQ基 頻訊號延遲電力計算部13或截波處理部17中的運算處理 時間。此外，加減法器23，24係分別由所延遲的iq訊號的 各成分I，Q中減去抵消訊號Ic,Qc，而輸出屬於訊號處理後 之IQ訊號的I〇ut,Qout。 瞬時電力P超過臨限値Pth的IQ基頻訊號因爲該減算 而被補正爲相當於臨限値Pth的瞬時電力的訊號。此外， 瞬時電力P爲臨限値Pth以下的IQ基頻訊號則未作補正而 直接輸出。 第5圖係顯示進行上述截波處理時之iq基頻訊號與臨 限値Pth的關係的IQ平面座標圖。 如該第5圖所示，本實施形態之峰値電力抑制電路9 之訊號處理係截斷IQ基頻訊號的瞬時電力P的外周側的截 波處理。爲此，電力放大電路5的電力放大器的pAPR降 低，因此電力放大器的電力效率會提升。 〔關於抵消用脈衝〕 第4圖係顯示抵消用脈衝S之生成方法的波形圖。 如第4圖所示’抵消用脈衝S係由將基本脈衝S a與輔 助脈衝Sb合成的合成脈衝所構成。 -16- 201212595 其中，在第4圖中，基本脈衝Sa的時間波形與頻譜顯 示於左上框內，輔助脈衝Sb的時間波形與頻譜顯示於左下 框內。此外，抵消用脈衝S的時間波形與頻譜顯示於右框 內。 上述基本脈衝 Sa係與專利文獻 3(日本特開 2004- 1 35078號公報）的情形相同，由將下行訊號的送訊所 使用的頻寬（以下有稱爲「使用頻寬」的情形）B所包含的複 數個（例如設爲N個）載波，振幅設爲1/N且相位設爲0，輸 入至前述IFFT部8所得的Sine波形所成者。此時，IFFT 部8的輸出僅出現實部I，虛部Q爲零。 如上所示，基本脈衝Sa係對IQ基頻訊號的頻帶B所 包含的複數個副載波，與其訊號的情形相同在IFFT部8進 行逆傅立葉轉換所得的實部I的波形（Sine波形）。 因此，基本脈衝Sa的頻帶係與使用頻寬B —致，即使 使用對超過臨限値Pth的IQ訊號的增量乘上基本脈衝a所 得的抵消訊號而將IQ訊號進行截波，在使用頻寬B的範圍 外亦不會發生不需要的頻率成分。 但是，在上述基本脈衝Sa中，由於僅使用送訊所使用 的頻帶B內的訊號成分，因此如第4圖之左上框內的時間 波形所示，無法將該時間軸上的脈衝寬度形成爲太細。 爲此，若僅採用上述基本脈衝Sa作爲抵消用脈衝S， 並以將其乘上增量M，AQ而求出的抵消訊號Ic，Qc來抵消 IQ基頻訊號時，抵消訊號Ic，Qc會干擾峰値時以後的訊號 -17- 201212595 波形而生成新的峰値波形，而會有無法確實地抑制IQ基頻 訊號的峰値電力的情形。 因此，本實施形態爲了即使使用頻寬B外的頻寬亦可 適當進行某一程度的峰値電力的抑制，除了基本脈衝Sa以 外，另外定義輔助脈衝Sb，並採用將該輔助脈衝Sb與基 本脈衝Sa合成者作爲抵消用脈衝S。 該輔助脈衝Sb係如第4圖之左下框內的時間波形所 示，由在基本脈衝Sa的峰値保持的時間內急遽上升，且由 接近於寬度比基本脈衝S a更細之非常細的得耳他函數 (D e 11 a f u n c t i ο η)的脈衝波形所構成。 因此，輔助脈衝Sb的頻譜爲包含使用頻寬Β之非常寬 的頻寬（以下有稱爲「廣頻寬」的情形）Bw。 如上所示，本實施形態之抵消用脈衝.S係由將上述輔 助脈衝Sb與習知的基本脈衝Sa合成的合成脈衝所構成， 因此如第4圖之右框內的頻譜所示，不僅IQ基頻訊號的頻 帶B而亦具有在該頻寬B之外的廣頻寬Bw的頻率成分。 〔各脈衝的峰値位準〕 此外，本實施形態若將基本脈衝Sa的峰値位準設爲 α，將輔助脈衝 Sb的峰値位準設爲 β，則以滿足 0.03 $ β/α$ 0.1的方式設定該等位準α，β的比率。以下針對 該理由進行說明。 如上所述，輔助脈衝Sb的頻率成分係遍及偏離IQ基 頻訊號之使用頻寬B的廣頻寬Bw，因此若由IQ基頻訊號 -18 - 201212595 減去使用包含輔助脈衝Sb的合成脈衝所求出的 Ic，Qc，則與遍及寬廣頻帶Bw而被施加雜訊爲相 因此，若未適當設定基本脈衝Sa與輔助脈復 準，會有使用頻寬B中的通信品質（EVM)惡化、 頻寬B的頻寬外發生高位準雜訊的可能性。 在此，例如在LTE中，被要求確保使用頻寬B 爲40dB ;此外，關於鄰近通道洩漏電力比，在電 要求確保60dB。 因此，合成輔助脈衝Sb而爲使用頻寬B所容 降低爲最大20dB，若將此換算爲電壓，即爲0.1 輔助脈衝Sb的峰値位準（電壓）的比率β/α係可容 爲0.1爲止。 另一方面，若輔助脈衝Sb的峰値位準過小， 法適當取消因抵消訊號Ic，Qc的減算所產生的新 之虞，但是若輔助脈衝Sb的峰値位準（電壓）的比 小爲0.03左右，則可明確取消新峰値波形。 由以上，關於各脈衝S1，S2的峰値位準ot，f 0.03$β/α$0.1的方式來設定該等位準α，β的比S 〔第1實施形態之效果〕 根據本實施形態之峰値電力抑制電路9，截 17係在該IQ基頻訊號減去對由IQ基頻訊號的S 的增量AI，AQ乘上不僅具有IQ基頻訊號的頻帶 率成分而亦具有該頻寬外的頻率成分的抵消用脈 抵消訊號 同。 i Sb的位 或在使用 中的EVM 波法上被 許的電力 。因此， 許至最大 則會有無 峰値波形 率β / α最 i，以滿足 每即可。 波處理部 ΐ限値Pth B內的頻 衝S(參照 -19- 201212595 第4圖）所得的抵消訊號I c，Q c。 0 因此，藉由影響頻寬外的頻率成分的抵消訊號Ic，Qc 的減算，可防止產生新的峰値波形，可更加確實地抑制IQ 基頻訊號的峰値電力。 〔第2實施形態〕 第7圖係第2實施形態之峰値電力抑制電路的功能方 塊圖。 如第7圖所示，本實施形態之峰値電力抑制電路9(第 7圖）與第1實施形態之峰値電力抑制電路9(第3圖）不同之 處在於另外具備有平均計算部33與臨限値更新部34。 以下，與第1實施形態共通的構成及功能在圖示中標 記相同元件符號並省略說明，並針對與第1實施形態的不 同之處作重點說明。 平均計算部3 3係取得送訊電力大幅變動所得之屬於 最小時間單位的OFDM符號的符號周期作爲計算IQ基頻訊 號的平均電力Pave的控制周期。 亦即，平均計算部33係由電力計算部13取得IQ基頻 訊號的瞬時電力P，將該瞬時電力P在上述符號周期內平 均化來計算每一個符號周期的IQ基頻訊號的平均電力 Pave，且將其輸出至臨限値更新部34。 臨限値更新部34係採用對由平均計算部33所取得的 每一個符號周期的平均電力Pave乘上預定倍率所得的値 作爲該符號周期中的臨限値Pth。例如，若將IQ基頻訊號 -20- 201212595 的峰値電力Ppeak與平均電力Pave的比率縮小爲6dB，則 上述預定倍率成爲2倍。 臨限値更新部3 4係如上所述按每一個符號周期計算 臨限値Pth而動態更新該臨限値Pth，並將該更新後的臨限 値Pth輸出至增量率計算部20與比較部21。 接著’比較部2 1係使用臨限値更新部3 4所取得的臨 限値P th ’判定電力計算部1 8所算出的瞬時電力P的大小， 當瞬時電力P超過更新後的臨限値Pth時，即對脈衝保持 部22發出抵消用脈衝S的輸出指令。 第8圖係顯示IQ基頻訊號的瞬時電力p與逐次更新的 臨限値Pth的時間變化的曲線圖。 如第8圖所示，在本實施形態中，峰値電力抑制電路 9中的截波處理所使用的臨限値pth係根據按每一個符號 周期（l/14ms)所算出的平均電力Pave而被逐次計算，且按 每一個該符號周期更新。 爲此’例如即使IQ基頻訊號的平均電力Pave對應於 移動終端機2所致之通話量的變動而發生變動，亦會經常 進行藉峰値電力抑制電路9之截波處理，因此可有效確保 因PAPR的減低所造成的功率放大器的電力效率的提升。 此外’根據本實施形態之峰値電力抑制電路9，由於 採用送訊電力可變動之屬於最小時間單位的OFDM的符號 周期作爲更新臨限値Pth的控制周期，因此亦有可正確且 迅速地更新臨限値Pth的優點。 -21 - 201212595 但是，與第1實施形態的同情形，LTE係資源區塊（參 照第6圖）爲使用者分配的最小單位，因此亦可採用屬於該 資源區塊之送訊周期的70FDM符號（1時槽）作爲更新臨限 値Pth的控制周期。 〔第3實施形態〕 第9圖係第3實施形態之無線通信系統的全體構成 圖。此外，第10圖係顯示此時之基地台裝置1之OFDM送 訊機3之主要部分的功能方塊圖。 在本實施形態中，亦採用根據LT E方式的無線通信系 統。該方式的基地台裝置1例如可以5 Μ Η z單位來設定下 行訊框的頻帶，對單元內的各移動終端機2傳送下行訊號 時，可按該每一個頻帶來變更送訊電力。 本實施形態之基地台裝置1係例示以2種頻帶Β 1，Β 2 傳送下行訊框的情形，頻率小者的第1頻寬Β 1的送訊電力 被設定爲較大，頻率大者的第2頻寬Β2的送訊電力被設定 爲較小。 因此，如第9圖中虛線所示，送訊電力較大之第1頻 寬Β1的下行訊號到達的通信區域Α1係比送訊電力較小之 第2頻寬Β2的下行訊號到達的通信區域Α2更爲遠方且範 圍更廣。 上述通信區域Α1，Α2重複的區域內移動終端機2可使 第1及第2頻寬Β1，Β2雙方進行通信，因此即使在通話量 較多的情形下，亦可確實地進行移動終端機2的通信。 -22- 201212595 本實施形態係假定基地台裝置1以2種頻帶B1，B2傳 送下行訊號的情形，因此如第1 0圖所示，副載波被包含在 第1頻寬B1的第1訊號I1，Q1、及副載波被包含在第2頻 寬B2的第2訊號I2，Q2會由IFFT部8被輸出。 該第1訊號II，Q1與第2訊號12, Q2係被輸入至後段 的訊號處理部（本實施形態之峰値電力抑制電路）9,在該處 理部9中被施行預定的訊號處理。 〔峰値電力抑制電路的構成〕 第1 1圖係第3實施形態之峰値電力抑制電路的功能方 塊圖。 如第1 1圖所示，本實施形態之峰値電力抑制電路9(第 11圖）與第1實施形態之峰値電力抑制電路9(第3圖）不同 之處在於另外具備電力計算部14,15與脈衝生成部16。 以下，與第1實施形態共通的構成及功能係在圖示中 標記相同元件符號並省略說明，針對與第1實施形態的不 同之處作重點說明。 其中，以下將第1訊號II，Q1與第2訊號12, Q2的合 成訊號僅稱爲「IQ基頻訊號」或「IQ訊號」。 此外，將第1訊號II，Q1的瞬時電力設爲P1，將第2 訊號12,Q2的瞬時電力設爲P2,將IQ基頻訊號的瞬時電力 設爲 P(= PI + P2)。 在本實施形態之峰値電力抑制電路9中，電力計算部 I4係計算由第1訊號I1，Q1的I成分（II)與Q成分（Q1)的 -23- 201212595 平方和所構成的第1訊號HQ1的瞬時電力ρι(=π2 + Q12)，電力計算部15係計算由第2訊號12, Q2的I成分（12) 與Q成分（Q2)的平方和所構成的第2訊號I2，Q2的瞬時電 力 P2( = I22 + Q22) 〇 〔脈衝生成部的構成〕 第12圖係脈衝生成部16的功能方塊圖。 該脈衝生成部16係對按每一個第1及第2頻寬B1,B2 所預先求出的合成脈衝 S1，S2，分別乘上毎一個該頻寬 B1，B2的平均電力的相對比率C1，C2而取總和來生成前述 抵消用脈衝S，而該脈衝生成部16具有比率計算部26、波 形記憶部2 7及乘加算部2 8。 其中，波形記憶部27係由記憶每一個頻帶B 1，B2的合 成脈衝S 1，S2的記憶體等記憶裝置所構成。該合成脈衝 S 1，S2係與第1實施形態的相同情形，將前述基本脈衝Sa 與輔助脈衝Sb(第4圖）合成。 但是，第1頻寬B1用的合成脈衝S1係對第1頻寬B1 所包含的複數個副載波，將由與送訊訊號的情形相同以 IFFT部8進行逆傅立葉轉換所得的實部I的波形所構成的 基本脈衝Sa與輔助脈衝Sb合成者。 此外，第2頻寬B2用的合成脈衝S2係對第2頻寬B2 所包含的複數個副載波，將由與送訊訊號的情形相同以 IFFT部8進行逆傅立葉轉換所得的實部I的波形所構成的 基本脈衝Sa與輔助脈衝Sb合成者。 -24 - 201212595 另一方面，比率計算部26分別被輸入電力計算部14 所算出的第1訊號II，Q1的瞬時電力P1、及電力計算部15 所算出的第2訊號12, Q2的瞬時電力P2。比率計算部26 係使用該等瞬時電力PI、P2，根據下式計算出每一個頻帶 B1，B2的平均電力的相對比率C1，C2。Qc = AQ χ S = { 1 - SQRT(Pth/P) } xQxS The delay unit 18, 19 located before the adder-subtracter 23, 24 delays the IQ fundamental signal delay power calculation unit 13 or the chopping processing unit 17 The processing time of the operation. Further, the adder-subtracters 23, 24 subtract the cancel signals Ic, Qc from the components I, Q of the delayed iq signal, respectively, and output I〇ut, Qout belonging to the IQ signal after the signal processing. The IQ fundamental frequency signal whose instantaneous power P exceeds the threshold 値Pth is corrected by the subtraction as a signal corresponding to the instantaneous power of the threshold 値Pth. In addition, the IQ fundamental frequency signal whose instantaneous power P is less than the threshold 値Pth is directly corrected without being corrected. Fig. 5 is an IQ plane coordinate diagram showing the relationship between the iq fundamental frequency signal and the threshold 値Pth when the above-described chopping processing is performed. As shown in Fig. 5, the signal processing of the peak power suppression circuit 9 of the present embodiment intercepts the outer peripheral side of the instantaneous power P of the IQ fundamental frequency signal. For this reason, the pAPR of the power amplifier of the power amplifying circuit 5 is lowered, so that the power efficiency of the power amplifier is improved. [Regarding the canceling pulse] Fig. 4 is a waveform diagram showing a method of generating the canceling pulse S. As shown in Fig. 4, the canceling pulse S is composed of a combined pulse combining the basic pulse S a and the auxiliary pulse Sb. -16- 201212595 Here, in Fig. 4, the time waveform and spectrum of the basic pulse Sa are displayed in the upper left frame, and the time waveform and spectrum of the auxiliary pulse Sb are displayed in the lower left frame. Further, the time waveform and spectrum of the canceling pulse S are displayed in the right frame. The basic pulse Sa is the same as the case of the patent document 3 (JP-A-2004-13535078), and the bandwidth used for the transmission of the downlink signal (hereinafter referred to as "use bandwidth") B The plurality of (for example, N) carriers included, the amplitude is 1/N, and the phase is set to 0, and the Sine waveform obtained by the IFFT unit 8 is input. At this time, the output of the IFFT section 8 shows only the real part I, and the imaginary part Q is zero. As described above, the basic pulse Sa is a waveform (Sine waveform) of the real part I obtained by inverse Fourier transform in the IFFT section 8 for a plurality of subcarriers included in the frequency band B of the IQ fundamental frequency signal as in the case of the signal. Therefore, the frequency band of the basic pulse Sa is consistent with the use bandwidth B, and the IQ signal is chopped even if the offset signal obtained by multiplying the basic pulse a by the increment of the IQ signal exceeding the threshold 値Pth is used. Unwanted frequency components do not occur outside the range of width B. However, in the basic pulse Sa, since only the signal component in the frequency band B used for the transmission is used, the pulse width on the time axis cannot be formed as shown by the time waveform in the upper left frame of FIG. too thin. Therefore, if only the basic pulse Sa is used as the canceling pulse S, and the offset signal Ic, Qc obtained by multiplying it by the increments M, AQ is used to cancel the IQ fundamental signal, the canceling signal Ic, Qc will be A new peak-to-peak waveform is generated by the signal -17-201212595 after the peak of the peak, and there is a case where the peak-to-peak power of the IQ fundamental signal cannot be reliably suppressed. Therefore, in the present embodiment, in order to suppress the peak-to-peak power of a certain degree even if the bandwidth outside the bandwidth B is used, the auxiliary pulse Sb is additionally defined in addition to the basic pulse Sa, and the auxiliary pulse Sb is basically used. The pulse Sa synthesizer is used as the canceling pulse S. The auxiliary pulse Sb is sharply rising in the time period held by the peak 基本 of the basic pulse Sa as shown by the time waveform in the lower left frame of FIG. 4, and is very finer than the width of the basic pulse Sa. A pulse waveform of the dear function (D e 11 afuncti ο η). Therefore, the spectrum of the auxiliary pulse Sb is a very wide bandwidth (hereinafter referred to as "wide bandwidth") Bw including the use bandwidth Β. As described above, the canceling pulse .S of the present embodiment is composed of a combined pulse in which the auxiliary pulse Sb is combined with the conventional basic pulse Sa. Therefore, as shown by the spectrum in the right frame of Fig. 4, not only IQ The frequency band B of the fundamental frequency signal also has a frequency component of the wide frequency bandwidth Bw outside the bandwidth B. [Peak level of each pulse] In the present embodiment, when the peak value of the basic pulse Sa is set to α, and the peak value of the auxiliary pulse Sb is set to β, the value of 0.03 $ β/α$ is satisfied. The mode of 0.1 sets the ratio of the levels α, β. The reason will be described below. As described above, the frequency component of the auxiliary pulse Sb is spread over the wide bandwidth Bw of the use bandwidth B of the IQ fundamental signal, so if the composite pulse including the auxiliary pulse Sb is subtracted from the IQ fundamental signal -18 - 201212595 The obtained Ic and Qc are in phase with the application of noise throughout the wide frequency band Bw. Therefore, if the basic pulse Sa and the auxiliary pulse registration are not properly set, the communication quality (EVM) in the used bandwidth B is deteriorated. The possibility of high level quasi-noise occurs outside the bandwidth of bandwidth B. Here, for example, in LTE, it is required to ensure that the use bandwidth B is 40 dB; moreover, with respect to the adjacent channel leakage power ratio, 60 dB is ensured in the electrical requirement. Therefore, the auxiliary pulse Sb is synthesized and reduced to a maximum of 20 dB for the use bandwidth B. If this is converted into a voltage, the ratio β/α of the peak 値 level (voltage) of the auxiliary pulse Sb can be 0.1. until. On the other hand, if the peak 値 level of the auxiliary pulse Sb is too small, the method appropriately cancels the new enthalpy generated by the subtraction of the canceling signals Ic, Qc, but if the ratio of the peak 値 level (voltage) of the auxiliary pulse Sb is small Approximately 0.03 or so, the new peak waveform can be explicitly cancelled. From the above, the ratio S of the levels α and β is set so that the peak value ot ot of the respective pulses S1 and S2 is f 0.03$β/α$0.1. [Effect of the first embodiment] According to the present embodiment The peak power suppression circuit 9 intercepts 17 the base frequency signal minus the incremental AI of the S of the IQ fundamental signal, and the AQ multiplied by the band rate component having not only the IQ fundamental signal but also the bandwidth. The offset of the outer frequency components is the same as the pulse cancellation signal. The bit of i Sb or the power applied to the EVM wave method in use. Therefore, if the maximum is reached, there will be no peak 値 waveform rate β / α most i, to meet each. The wave processing unit is limited to the offset signal I c, Q c obtained by the frequency S in the Pth B (refer to -19-201212595, Fig. 4). Therefore, by subtracting the offset signals Ic and Qc which affect the frequency components outside the bandwidth, it is possible to prevent a new peak-to-peak waveform from being generated, and it is possible to more reliably suppress the peak-to-peak power of the IQ fundamental signal. [Second Embodiment] Fig. 7 is a functional block diagram of a peak-to-peak power suppression circuit according to a second embodiment. As shown in Fig. 7, the peak power suppression circuit 9 (Fig. 7) of the present embodiment is different from the peak power suppression circuit 9 (Fig. 3) of the first embodiment in that an average calculation unit 33 is provided. And the threshold update unit 34. In the following, the same components and functions as those in the first embodiment are denoted by the same reference numerals, and their description is omitted, and the differences from the first embodiment will be mainly described. The average calculation unit 3 3 obtains a symbol period of the OFDM symbol belonging to the minimum time unit obtained by greatly varying the transmission power as a control period for calculating the average power Pave of the IQ fundamental signal. That is, the average calculating unit 33 obtains the instantaneous power P of the IQ fundamental frequency signal from the power calculating unit 13, and averages the instantaneous power P in the symbol period to calculate the average power Pave of the IQ fundamental frequency signal for each symbol period. And output it to the threshold update unit 34. The threshold update unit 34 uses 値 which multiplies the average power Pave for each symbol period acquired by the average calculation unit 33 by a predetermined magnification as the threshold 値Pth in the symbol period. For example, when the ratio of the peak power Ppeak of the IQ fundamental frequency signal -20-201212595 to the average power Pave is reduced to 6 dB, the predetermined magnification is doubled. The threshold update unit 34 calculates the threshold thPth for each symbol period as described above, dynamically updates the threshold 値Pth, and outputs the updated threshold 値Pth to the increment rate calculation unit 20 and compares Department 21. Next, the comparison unit 2 1 determines the magnitude of the instantaneous power P calculated by the power calculation unit 18 using the threshold 値P th ' obtained by the threshold update unit 34, and when the instantaneous power P exceeds the updated threshold 値At the time of Pth, the pulse holding unit 22 issues an output command for the canceling pulse S. Figure 8 is a graph showing the temporal variation of the instantaneous power p of the IQ fundamental frequency signal and the successively updated threshold 値Pth. As shown in Fig. 8, in the present embodiment, the threshold 値pth used for the occlusion processing in the peak power suppression circuit 9 is based on the average power Pave calculated for each symbol period (l/14 ms). It is calculated successively and updated every one symbol period. For this reason, for example, even if the average power Pave of the IQ fundamental signal changes in accordance with the fluctuation of the amount of the call caused by the mobile terminal 2, the interception processing by the peak power suppression circuit 9 is often performed, so that it can be effectively ensured. The power efficiency of the power amplifier is increased due to the reduction in PAPR. Further, according to the peak chirp power suppression circuit 9 of the present embodiment, since the symbol period of the OFDM belonging to the minimum time unit which can be varied by the transmission power is used as the control period of the update threshold 値Pth, it is also possible to update correctly and quickly. The advantage of the threshold 値Pth. -21 - 201212595 However, in the same manner as the first embodiment, the LTE-based resource block (see Fig. 6) is the smallest unit allocated to the user, and therefore the 70FDM symbol belonging to the transmission period of the resource block can also be used. (1 hour slot) as the control period for updating the threshold 値Pth. [Third Embodiment] Fig. 9 is a view showing the overall configuration of a radio communication system according to a third embodiment. Further, Fig. 10 is a functional block diagram showing the main part of the OFDM transmitter 3 of the base station apparatus 1 at this time. In the present embodiment, a wireless communication system according to the LT E method is also employed. The base station apparatus 1 of this type can set the frequency band of the downlink frame, for example, in units of 5 Η Η z, and when transmitting the downlink signal to each mobile terminal 2 in the unit, the transmission power can be changed for each frequency band. In the base station apparatus 1 of the present embodiment, the case where the downlink frame is transmitted in two types of bands Β 1, Β 2 is exemplified, and the transmission power of the first bandwidth Β 1 of the smaller frequency is set to be larger, and the frequency is larger. The transmission power of the second bandwidth Β2 is set to be small. Therefore, as indicated by the broken line in FIG. 9, the communication area Α1 in which the downlink signal of the first bandwidth 较大1 having the larger transmission power is larger is the communication area in which the downlink signal of the second bandwidth 较小2 having the smaller transmission power arrives. Α2 is farther and wider. In the area where the communication area Α1, Α2 is repeated, the mobile terminal 2 can communicate with both the first and second bandwidths Β1 and Β2. Therefore, even when the amount of calls is large, the mobile terminal 2 can be surely performed. Communication. -22-201212595 In the present embodiment, the base station apparatus 1 assumes that the downlink signal is transmitted in two types of bands B1 and B2. Therefore, as shown in FIG. 1, the subcarrier is included in the first signal I1 of the first bandwidth B1. Q1 and subcarriers are included in the second signal I2 of the second bandwidth B2, and Q2 is outputted by the IFFT unit 8. The first signal II, Q1, and the second signal 12, Q2 are input to the signal processing unit (the peak power suppression circuit of the present embodiment) 9 in the subsequent stage, and the processing unit 9 performs predetermined signal processing. [Configuration of the peak power suppression circuit] Fig. 1 is a functional block diagram of the peak power suppression circuit of the third embodiment. As shown in FIG. 1 , the peak power suppression circuit 9 (FIG. 11) of the present embodiment is different from the peak power suppression circuit 9 (third diagram) of the first embodiment in that the power calculation unit 14 is separately provided. 15 and the pulse generating unit 16. In the following, the same components and functions as those in the first embodiment are denoted by the same reference numerals, and their description is omitted, and the differences from the first embodiment will be mainly described. In the following, the synthesis signal of the first signal II, Q1 and the second signal 12, Q2 is simply referred to as "IQ fundamental frequency signal" or "IQ signal". Further, the instantaneous power of the first signal II, Q1 is P1, the instantaneous power of the second signal 12, Q2 is P2, and the instantaneous power of the IQ fundamental signal is P (= PI + P2). In the peak power suppression circuit 9 of the present embodiment, the power calculation unit I4 calculates the first of the sum of the -23-201212595 sum of the I component (II) of the first signal I1 and Q1 and the Q component (Q1). The instantaneous power ρι (= π2 + Q12) of the signal HQ1, the power calculation unit 15 calculates the second signal I2, Q2 composed of the sum of the squares of the I component (12) and the Q component (Q2) of the second signal 12, Q2. Instantaneous electric power P2 (= I22 + Q22) 〇 [Configuration of pulse generating unit] Fig. 12 is a functional block diagram of the pulse generating unit 16. The pulse generation unit 16 multiplies the composite pulses S1 and S2 obtained in advance for each of the first and second bandwidths B1 and B2 by the relative ratio C1 of the average power of the bandwidths B1 and B2, respectively. The canceling pulse S is generated by taking the sum of C2, and the pulse generating unit 16 includes a ratio calculating unit 26, a waveform memory unit 27, and a multiplying and adding unit 28. The waveform memory unit 27 is constituted by a memory device such as a memory that stores the synthesis pulses S1 and S2 of each of the frequency bands B1 and B2. The combined pulses S1, S2 are combined with the auxiliary pulse Sb (Fig. 4) in the same manner as in the first embodiment. However, the composite pulse S1 for the first bandwidth B1 is a waveform of the real part I obtained by inverse Fourier transform of the IFFT unit 8 in the same plurality of subcarriers included in the first bandwidth B1 as in the case of the transmission signal. The basic pulse Sa and the auxiliary pulse Sb are combined. Further, the combined pulse S2 for the second bandwidth B2 is a waveform of the real part I obtained by inverse Fourier transform of the IFFT portion 8 in the same plurality of subcarriers included in the second bandwidth B2 as in the case of the transmission signal. The basic pulse Sa and the auxiliary pulse Sb are combined. -24 - 201212595 On the other hand, the ratio calculation unit 26 is input to the instantaneous power P1 of the first signal II, Q1 and the instantaneous power of the second signal 12, Q2 calculated by the power calculation unit 15 calculated by the power calculation unit 14, respectively. P2. The ratio calculating unit 26 calculates the relative ratios C1 and C2 of the average power of each of the frequency bands B1 and B2 based on the instantaneous powers PI and P2.
Cl = Σ/· Ρ1/(Σ/ Ρ1+ Ι/' Ρ2) C2= Σ，Ρ2/(Σ，Ρ1 + Σ，Ρ2) 如上述計算式所示，每一個頻帶Β1，Β2的平均電力的 相對比率C1，C2係以預定的取樣周期累積每一個該頻帶 B1，B2的瞬時電力PI、P2的平方根/*pl、入p2，並將該累 積値，pl、p2除以各頻帶B1，B2之累積値的總和（Σ入 pl+X/· P2)而求出。 比率計算部26係取得送訊電力可大幅變動之屬於最 小時間單位OFDM符號的符號周期作爲控制周期，在該符 號周期內執行上述相對比率C1,C2的計算。 如此一來，可在IQ基頻訊號的平均電力不太會變動的 安定狀態下計算相對比率C 1 , C 2，因而具有可得到正確的 相對比率Cl，C2的效果。 但是，LTE係資源區塊（參照第6圖）爲使用者分配的最 小單位，因此亦可採用屬於該資源區塊之送訊周期的 70FDM符號（1時槽）作爲計算相對比率ci，C2時的控制周 期。 乘加算部28係包含2個乘法器2 9，30及1個加算器 -25- 201212595 31。其中，乘法器29係對與第1頻寬B1對應的相對比率 C1乘以該頻寬B1用的合成脈衝S1’乘法器30係對與第2 頻寬B2對應的相對比率C2乘以該頻寬B2用的合成脈衝 C2 〇 此外’加算器31係加上各乘法器29,30.的乘算結果而 生成抵消用脈衝S，且將該脈衝S輸出至截波處理部17的 脈衝保持部22。亦即，乘加算器16係根據下式而生成抵 消用脈衝S。 S= ClxSl+ C2xS2 本實施形態之乘加算部2 8係將在比率計算部2 6所算 出的相對比率C 1，C2與預定的臨限値作比較來判定其變 動，僅在變動成相對比率C 1，C2超過臨限値的程度的情形 下，執行使用該變動後的相對比率C 1，C2的乘算及總和， 將其結果所生成的抵消用脈衝S輸出至脈衝保持部22。 因此，只要相對比率Cl,c 2在一定程度下不會變動， 乘加算器28不會執行乘算及總和，而脈衝保持部22維持 以前的抵消周脈衝S。因此’與在每次愚直地生成抵消用 脈衝S的情形相比，可減低電路的運算負荷。 〔第3實施形態之效果〕 上述抵消用脈衝s係將對與第1頻寬B1對應的第1 訊號I1，Q1的平均電力的相對比率C1乘以該頻寬B1用的 合成脈衝S1者、及對與第2頻寬B2對應的第2訊號12, Q2 的平均電力的相對比率C2乘以該頻寬B2用的合成脈衝S2 -26- 201212595 者進行加算者。 因此，即使從原本的IQ基頻訊號減去將上述抵消用脈 衝S乘以增量所得的抵消訊號Ic，Qc，亦與每一個第 1及第2頻寬B1，B2的平均電力對應而抵消IQ基頻訊號的 振幅。 因此，根據本實施形態之峰値電力抑制電路9，即使 在第1及第2頻寬B1，B 2中的平均電力有差異的情形下， 亦不會有平均電力較小者的頻寬B2的送訊電力因抵消訊 號Ic，Qc的減算而降低爲所需以上的情形，即使爲平均電 力按每一個頻帶B1，B2而異的IQ基頻訊號的情形下，亦不 會使SNR惡化，而可適當進行截波處理。 〔第4實施形態〕 第13圖係本發明之第4實施形態之無線通信系統的全 體構成圖。 如第1 3圖所示，在本實施形態之無線通信系統中，在 基地台裝置 1 透過 C P RI (C o m m ο η P u b 1 i c R a d i ο I n t e r f a c e) 連接有 RRH(Remote Radio H e ad) 3 6 > 在該 RRH36 設有第 14圖所示之第4實施形態之峰値電力抑制電路9與前述電 力放大電路5。 此外，本實施形態，基地台裝置1係將在與RRH3 6之 間建立同步的同步訊號38，透過光纖而送出至RRH36，該 同步訊號38係由與OFDM的符號周期同步的lms周期的 時脈訊號所構成。 -27- 201212595 如第1 4圖所示，本實施形態之峰値電力抑制電路9係 設有一周期生成部37，而上述同步訊號38被輸入此周期 生成部37。 該周期生成部37係由從屬於外部裝置的基地台裝置1 所取得的同步訊號38生成符號周期，將所生成的符號周期 輸出至脈衝生成部16與平均計算部33。其中，由於其他 構成與第2實施形態（第7圖）的峰値電力抑制電路9相同， 因此在第11圖中標記與第7圖相同的元件符號，且省略詳 細說明。 如上所述，本實施形態係由基地台裝置1取得與符號 周期同步的同步訊號38，並根據該同步訊號38來生成符 號周期’因此在RRH36亦可搭載本發明之峰値電力抑制電 路9。 〔基本脈衝的變化〕 第15圖係顯示基本脈衝Sa的變化的時域的曲線圖， 在第15圖中’（a)爲Sine波形、（b)爲卻比雪夫（ChebyShev) 波形、（c)爲泰勒（Taylor)波形。 該等波形在數學上係可全部以下式（1)表示，若爲sinc 波形則爲a n = η p ⑴ 州)=飢卜(ί)2Cl = Σ/· Ρ1/(Σ/ Ρ1+ Ι/' Ρ2) C2= Σ,Ρ2/(Σ,Ρ1 + Σ,Ρ2) As shown in the above formula, the relative ratio of the average power of each band Β1, Β2 C1, C2 accumulates the instantaneous power PI, the square root of the B2/*pl, enters p2 in a predetermined sampling period, and divides the accumulated 値, pl, p2 by the accumulation of each frequency band B1, B2. Find the sum of 値 (into pl+X/· P2). The ratio calculating unit 26 obtains a symbol period belonging to the minimum time unit OFDM symbol in which the transmission power can be largely changed as a control period, and performs calculation of the relative ratios C1 and C2 in the symbol period. In this way, the relative ratios C 1 , C 2 can be calculated in a stable state in which the average power of the IQ fundamental frequency signal does not fluctuate, and thus the effect of obtaining the correct relative ratios C1, C2 can be obtained. However, the LTE-based resource block (refer to FIG. 6) is the smallest unit allocated to the user. Therefore, the 70FDM symbol (1 time slot) belonging to the transmission period of the resource block can also be used as the relative ratio ci, C2 when calculating the relative ratio ci, C2. Control cycle. The multiplying and adding unit 28 includes two multipliers 2 9, 30 and one adder -25 - 201212595 31. The multiplier 29 multiplies the relative ratio C1 for the first bandwidth B1 by the composite pulse S1' for the bandwidth B1 by multiplying the relative ratio C2 corresponding to the second bandwidth B2 by the frequency. In addition to the multiplication result of each of the multipliers 29, 30, the summation counter 31 generates a canceling pulse S, and outputs the pulse S to the pulse holding portion of the chopping processing unit 17. twenty two. That is, the multiplier adder 16 generates the canceling pulse S based on the following equation. S = ClxSl + C2xS2 The multiplication and addition unit 28 of the present embodiment compares the relative ratios C 1, C2 calculated by the ratio calculating unit 26 with a predetermined threshold 来 to determine the fluctuation, and only changes to the relative ratio C. When the value of C2 exceeds the limit, the multiplication and sum of the relative ratios C1 and C2 after the change are performed, and the canceling pulse S generated as a result is output to the pulse holding unit 22. Therefore, as long as the relative ratio C1, c2 does not fluctuate to some extent, the multiplier adder 28 does not perform the multiplication and the sum, and the pulse holding unit 22 maintains the previous canceling cycle pulse S. Therefore, the calculation load of the circuit can be reduced as compared with the case where the canceling pulse S is generated in a foolish manner. [Effects of the third embodiment] The canceling pulse s is obtained by multiplying the relative ratio C1 of the average power of the first signals I1 and Q1 corresponding to the first bandwidth B1 by the combined pulse S1 for the bandwidth B1. And the addition ratio C2 of the average power of the second signal 12 and Q2 corresponding to the second bandwidth B2 is multiplied by the combined pulse S2-26-201212595 for the bandwidth B2. Therefore, even if the offset signal Ic, Qc obtained by multiplying the offset pulse S by the increment is subtracted from the original IQ fundamental signal, it is offset by the average power of each of the first and second bandwidths B1, B2. The amplitude of the IQ baseband signal. Therefore, according to the peak-to-peak power suppression circuit 9 of the present embodiment, even when the average power in the first and second bandwidths B1 and B2 is different, there is no bandwidth B2 in which the average power is smaller. The transmission power is reduced to more than necessary due to the subtraction of the cancellation signals Ic, Qc, and even if the average power is in the IQ fundamental signal of each frequency band B1, B2, the SNR is not deteriorated. The chopping process can be performed as appropriate. [Fourth Embodiment] Fig. 13 is a view showing the overall configuration of a radio communication system according to a fourth embodiment of the present invention. As shown in FIG. 3, in the radio communication system according to the present embodiment, the base station apparatus 1 is connected to the RRH (Remote Radio H e ad) via CP RI (C omm ο P ub 1 ic R adi ο I nterface). 3 6 > The peak power suppression circuit 9 of the fourth embodiment shown in Fig. 14 and the power amplifier circuit 5 are provided in the RRH 36. Further, in the present embodiment, the base station apparatus 1 transmits a synchronization signal 38 synchronized with the RRH 36 to the RRH 36 through the optical fiber, and the synchronization signal 38 is a clock of the lms period synchronized with the symbol period of the OFDM. The signal consists of. -27-201212595 As shown in Fig. 14, the peak power suppression circuit 9 of the present embodiment is provided with a period generating unit 37, and the synchronization signal 38 is input to the period generating unit 37. The cycle generation unit 37 generates a symbol period from the synchronization signal 38 acquired by the base station device 1 belonging to the external device, and outputs the generated symbol cycle to the pulse generation unit 16 and the average calculation unit 33. The other configuration is the same as that of the peak power suppression circuit 9 of the second embodiment (Fig. 7). Therefore, the same reference numerals are given to elements in Fig. 11 and the detailed description thereof will be omitted. As described above, in the present embodiment, the base station device 1 acquires the synchronization signal 38 synchronized with the symbol period, and generates the symbol period based on the synchronization signal 38. Therefore, the peak power suppression circuit 9 of the present invention can be mounted on the RRH 36. [Basic Pulse Change] Fig. 15 is a time-domain graph showing the change of the basic pulse Sa. In Fig. 15, '(a) is a Sine waveform, (b) is a Cheby Shev waveform, (c) ) is the Taylor waveform. These waveforms can all be represented by the following formula (1) mathematically, and if they are sinc waveforms, a n = η p (1) state) = hunger (ί) 2
若爲 Sine，則 an=nK -28- 201212595 在此，若以包含絕對値的最大値且振幅値成爲零爲止 的區間（第15圖以影線所示的區間）爲主波瓣區間，則Sine 波形之情況因爲旁波瓣的振幅變得較大，無法過度提升主 波瓣區間的能量局部率。 相對於此，卻比雪夫波形係藉由調整構成爲振幅値=0 之X的解的數列an的値，可減小旁波瓣的振幅，但是此情 況係振幅變得不會衰減。 因此，泰勒波形係將數列an之起始的數點（例如a 1與 a2)的値使用卻比雪夫波形者，並使用Sine波形者作爲其 之後的點的値，藉此達成旁波瓣的振幅抑制與衰減特性兩 者。 因此，若比較相對定義基本脈衝Sa的預定時間區間T 中的全能量（振幅的平方）的主波瓣區間的能量局部率，Sine 波形的情形係91%，卻比雪夫波形的情形係93%，泰勒波 形的情形係約9 5 %，泰勒波形爲最有利。 其中，上述預定的時間區間T係被記錄在記憶體上的 波形的試樣時間，且爲與上限數的試樣點對應的時間。例 如，LTE之情況係1符號周期（l/14ms)所包含的試樣數爲 2048 ，因此若假設在時域進行 4 倍的超取樣 (oversampling)，則定義基本脈衝Sa之波形所需試樣點的 上限數成爲2048x4 = 8192個。 若將可在本發明中使用的基本脈衝Sa，以對主波瓣區 間之預定時間區間T的能量局部率的數値範圍特定時，該 -29- 201212595 能量局部率較佳爲8 5 %〜9 9 %。 其理由爲若能量局部率爲100%時，則基本脈衝Sa成 爲脈衝（得耳他函數），變成無法適用於具有頻帶限制的本 發明中，若局部率未達8 5 %，則脈衝形狀會過於鈍化而變 得無法使用。 基於以上，若列舉本發明所使用的基本脈衝Sa的技術 特徵，則如下所示。 特徵1 :基本脈衝S a可由相對預定的時間區間.T(例如 1符號周期）中的全能量（振幅的平方）的主波瓣區間的能量 局部率爲85%〜99%的波形所構成。 特徵2 :若以數學式記述基本脈衝S a，則由在時域中 具有對稱性之以前述式（1 )所表示的波形所構成。 特徵3:更具體而言，基本脈衝Sa係由Sine波形、卻 比雪夫波形或泰勒波形所構成。其中，Sine波形係將頻寬 內的複數個載波，由振幅相同且將相位設爲零而作逆傅立 葉轉換所得的實部（I訊號）的波形所構成。 〔其他變形例〕 本次所揭示之實施形態爲例示而並非爲具限制性者。 本發明之權利範圍係藉申請專利範圍予以表示，包含與申 請專利範圍的構成等效的範圍內的所有變更。 例如，上述第2實施形態係例示基地台裝置1使用2 個頻帶B1，B2的情形，但是即使在使用2個以上的頻帶的 情形下亦可構成本發明之峰値電力抑制電路9。 -30- 201212595 此外，本發明之峰値電力抑制電路9不僅LTE方式， 亦可採用於依據W-C DMA方式的通信裝置。 該W-CD Μ A方式係藉由封閉迴路送訊電力控制來控制 基地台裝置1的送訊電力，該控制周期爲送訊控制的最小 時間單位。具體而言，該控制周期係1無線訊框周期1 Oms 的 15 分之 1(=約 〇.667ms)。 因此，將本發明之峰値電力抑制電路9使用在W-CDMA 方式的送訊機係採用封閉迴路送訊電力控制的控制周期來 作爲相對比率C1，C2的計算或更新臨限値Pth時的控制周 期即可。 此外，上述實施形態係例示進行根據PC-CFR之截波 處理的峰値電力抑制電路9，但是本發明亦可適用於進行 根據NS-CFR的截波處理的峰値電力抑制電路9。 【圖式簡單說明】 第1圖係第1實施形態之無線通信系統的全體構成圖。 第2圖係顯示基地台裝置之OFDM送訊機之主要部分 的功能方塊圖。 第3圖係第1實施形態之峰値電力抑制電路的功能方 塊圖。 第4圖係顯示抵消用脈衝之生成方法的波形圖。 第5圖係顯示IQ基頻訊號與臨限値的關係的IQ平面 的座標圖。 第6圖係LTE的下行訊框的訊框構成圖。 -31- 201212595 第7圖係第2實施形態之峰値電力抑制電路的功能方 塊圖。 第8圖係顯示IQ基頻訊號的瞬時電力與逐次更新的臨 限値的時間變化的曲線圖。 第9圖係第3實施形態之無線通信系統的全體構成圖。 第10圖係顯示基地台裝置之OFDM送訊機的主要部分 的功能方塊圖。 第1 1圖係第3實施形態之峰値電力抑制電路的功能方 塊圖。 第1 2圖係脈衝生成部的功能方塊圖。 第1 3圖係第4實施形態之無線通信系統的全體構成 圖。 第1 4圖係第4實施形態之峰値電力抑制電路的功能方 塊圖。 · 第15圖係顯示基本脈衝之變化的時域的曲線圖。 【主要元件符號說明】 V 基地台裝置 2 移動終端機 3 送訊機 4 送訊用處理器 5 電力放大電路 6 S/P轉換部 7 映射部 -32- 201212595 8 IF F T ( c ο n v e r s e Fast Fourier Transform ：高速 傅立葉逆轉換）部 9 訊 號 處 理 部 (峰値電力抑制電路） 10 正 交 壬田 變 部 13 電 力 計 算 部 14 電 力 計 算 部 15 電 力 計 算 部 16 脈 衝 生 成 部 17 截 波 處. 理 部 18,19 延 遲 部 20 增 量 率 計 算 部 2 1 比 較 部 22 脈 衝 保 持 部 23,24 加 減 法 器 26 比 率 計 算 部 27 波 形 記 憶 部 28 乘 加 算 部 3 3 平 均 計 算 部 34 臨 限 値 更 新 部 37 周 期 生 成 部 3 8 同 步 訊 號 ΔΙ 增 量 △Q 增 量 -33- 201212595In the case of Sine, an=nK -28-201212595 Here, if the interval including the maximum 値 of the absolute 値 and the amplitude 値 becomes zero (the interval indicated by hatching in Fig. 15) is the main lobe section, In the case of the Sine waveform, since the amplitude of the side lobes becomes large, the energy local rate of the main lobe interval cannot be excessively increased. On the other hand, the Cheby wave waveform can reduce the amplitude of the side lobes by adjusting the 数 of the sequence an of the solution of the amplitude 値 = 0, but in this case, the amplitude is not attenuated. Therefore, the Taylor waveform uses the number of points at the beginning of the sequence an (for example, a 1 and a2 ) to be compared to the Cheval waveform, and uses the Sine waveform as the 値 of the point after it, thereby achieving the side lobes. Both amplitude suppression and attenuation characteristics. Therefore, if the energy local rate of the main lobe section of the total energy (square of the amplitude) in the predetermined time interval T of the basic pulse Sa is compared, the Sine waveform is 91%, but the case of the Schiff waveform is 93%. The Taylor waveform is about 95%, and the Taylor waveform is the most favorable. Here, the predetermined time interval T is a sample time of a waveform recorded on the memory, and is a time corresponding to the sample point of the upper limit. For example, in the case of LTE, the number of samples included in the 1-symbol period (l/14ms) is 2048, so if it is assumed that 4 times oversampling is performed in the time domain, the sample required to define the waveform of the basic pulse Sa is required. The upper limit of the point becomes 2048x4 = 8192. If the basic pulse Sa which can be used in the present invention is specified in the range of the number of energy local rates for the predetermined time interval T of the main lobe section, the energy local rate of the -29-201212595 is preferably 85%. 9 9 %. The reason is that if the local rate of energy is 100%, the basic pulse Sa becomes a pulse (a function of the ear), and it cannot be applied to the present invention having a band limitation. If the local rate is less than 85 %, the pulse shape will be It is too passivated and becomes unusable. Based on the above, the technical characteristics of the basic pulse Sa used in the present invention are as follows. Feature 1: The basic pulse S a can be composed of a waveform having a local rate of 85% to 99% of the main lobe section of the total energy (square of the amplitude) in a predetermined time interval .T (for example, 1 symbol period). Feature 2: When the basic pulse S a is described in a mathematical expression, it is composed of a waveform represented by the above formula (1) having symmetry in the time domain. Feature 3: More specifically, the basic pulse Sa is composed of a Sine waveform, a Chebysh waveform or a Taylor waveform. Among them, the Sine waveform is composed of a plurality of carriers in the bandwidth, which are composed of real-world (I-signal) waveforms obtained by inverse Fourier transform with the same amplitude and zero phase. [Other Modifications] The embodiments disclosed herein are illustrative and not restrictive. The scope of the invention is expressed by the scope of the claims, and all modifications within the scope equivalent to the composition of the claims. For example, the second embodiment is a case where the base station device 1 uses two frequency bands B1 and B2. However, the peak power suppression circuit 9 of the present invention can be configured even when two or more frequency bands are used. -30- 201212595 Further, the peak power suppression circuit 9 of the present invention can be applied not only to the LTE system but also to the communication device according to the W-C DMA method. The W-CD Μ A mode controls the transmission power of the base station device 1 by closed loop transmission power control, which is the minimum time unit of the transmission control. Specifically, the control period is 1/15 (= about 667 ms) of the radio frame period 1 Oms. Therefore, the peak power suppression circuit 9 of the present invention uses the control cycle of the closed loop communication power control in the W-CDMA type transmission machine as the relative ratio C1, C2 when calculating or updating the threshold 値Pth The control cycle is OK. Further, in the above embodiment, the peak-to-peak power suppression circuit 9 according to the cutoff processing of the PC-CFR is exemplified, but the present invention is also applicable to the peak-to-peak power suppression circuit 9 that performs the cutoff processing according to the NS-CFR. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a view showing the overall configuration of a wireless communication system according to a first embodiment. Fig. 2 is a functional block diagram showing the main part of the OFDM transmitter of the base station apparatus. Fig. 3 is a functional block diagram of the peak power suppression circuit of the first embodiment. Fig. 4 is a waveform diagram showing a method of generating a canceling pulse. Figure 5 is a graph showing the IQ plane of the relationship between the IQ fundamental signal and the threshold 値. Figure 6 is a block diagram of the downlink frame of LTE. -31-201212595 Fig. 7 is a functional block diagram of the peak power suppression circuit of the second embodiment. Figure 8 is a graph showing the temporal variation of the instantaneous power of the IQ fundamental signal and the time 値 of the successive update. Fig. 9 is a view showing the overall configuration of a wireless communication system according to a third embodiment. Fig. 10 is a functional block diagram showing the main part of the OFDM transmitter of the base station apparatus. Fig. 1 is a functional block diagram of the peak power suppression circuit of the third embodiment. Fig. 12 is a functional block diagram of the pulse generation unit. Fig. 13 is a view showing the overall configuration of a wireless communication system according to a fourth embodiment. Fig. 14 is a functional block diagram of the peak power suppression circuit of the fourth embodiment. • Figure 15 is a graph showing the time domain of changes in the fundamental pulse. [Description of main component symbols] V Base station device 2 Mobile terminal device 3 Transmitter 4 Transmitter processor 5 Power amplifier circuit 6 S/P converter unit 7 Mapping unit -32- 201212595 8 IF FT (c ο nverse Fast Fourier Transform: fast Fourier transform (transformation) unit 9 signal processing unit (peak power suppression circuit) 10 orthogonal field converter unit 13 power calculation unit 14 power calculation unit 15 power calculation unit 16 pulse generation unit 17 interception unit 19 Delay unit 20 Increment rate calculation unit 2 1 Comparison unit 22 Pulse hold unit 23, 24 Adder and subtractor 26 Ratio calculation unit 27 Waveform memory unit 28 Multiply-add unit 3 3 Average calculation unit 34 Threshold update unit 37 Period generation Part 3 8 Synchronization signal ΔΙ Incremental △ Q Increment -33- 201212595
Ic 抵 消 訊 號 Qc 抵 消 訊 Pc& Wl s 抵 消 用 脈衝 Sa 基 本 脈 衝 Sb 輔 助 脈 衝 -34Ic cancel signal Qc offset Pc& Wl s cancel pulse Sa basic pulse Sb auxiliary pulse -34